EP3455097A1

EP3455097A1 - Method for controlling voice-controlled operating interfaces in motor vehicles and device for performing the method

Info

Publication number: EP3455097A1
Application number: EP17713624.9A
Authority: EP
Inventors: David Spieler
Original assignee: Robert Bosch GmbH
Current assignee: Robert Bosch GmbH
Priority date: 2016-05-10
Filing date: 2017-03-24
Publication date: 2019-03-20
Anticipated expiration: 2037-03-24
Also published as: CN109153329A; EP3455097B1; DE102016207975A1; WO2017194241A1

Abstract

The invention relates to a method for controlling voice-controlled operating interfaces (12) in motor vehicles, comprising the following steps: a) determining an expected position of a motor vehicle within a predefined future time interval (T), b) determining static and/or dynamic stress sources at the expected position, c) assigning at least one stress value to at least one of the determined static and/or dynamic stress sources, d) determining an accumulated stress value from the at least one stress value and/or adopting at least one stress value as the accumulated stress value, e) recognizing a mental stress situation of a user of a voice-controlled operating interface (12) in the predefined future time interval (T), wherein the mental stress situation is recognized if the accumulated stress value at the expected position and/or in the predefined future time interval exceeds a predefined threshold value, and f) adapting an interaction between a voice-controlled operating interface (12) and a user if a stress situation has been recognized.

Description

Method for controlling voice-controlled operating interfaces in

Motor vehicles and apparatus for carrying out the method

description

The invention relates to a method for controlling voice-controlled operating interfaces in motor vehicles and to an apparatus for carrying out the method.

Voice-controlled operating interfaces are used, for example, for controlling navigation systems in motor vehicles. However, voice-controlled user interfaces can also be used for further functions of the

Motor vehicle controlled.

State of the art

From JP 2009-266100 A a device is known, which recognizes the current mental state of the driver, for example, his tiredness, based on track position data.

US 5,862,511 A discloses a vehicle navigation system which estimates a future vehicle position from the current position using speed and heading data.

A problem known voice controlled user interfaces is the mental load of the user and the associated distraction of the

Main events, such as road traffic. To endanger the user and other road users it often happens when an interaction with the voice-controlled user interface at the same time to critical driving situations, such as turning at intersections or overtaking maneuvers to take place. However, the known systems do not try to predict the problems of voice-controlled operating interfaces or to prevent them actively and preventively. In particular, the data used is not associated with any potential mental stress on the user.

Disclosure of the invention

The present invention is based on the object, a method and an apparatus for reducing the distraction of a user by

provide voice-controlled operator interfaces in road traffic.

The object underlying the invention is achieved by providing a method for controlling voice-controlled operating interfaces in motor vehicles, comprising the steps of: a) determining an expected position of a motor vehicle within a predetermined future time interval, b) determining static and / or dynamic load sources at the expected position, c) assigning at least one load value to at least one of the determined static and / or dynamic load sources, d) determining an accumulated load value from the at least one load value and / or applying at least one

E) detecting a mental stress situation of a user of a voice-operated user interface in the predetermined future time interval, wherein the mental stress situation is detected when the accumulated load value at the expected position and / or in the predetermined future time interval exceeds a predetermined threshold , and f) adjusting an interaction between a voice-controlled

Control interface and a user when a load situation has been detected, the adjustment of a time shift and / or a

Suspension and / or termination of a voice input and / or a

Speech output and / or a total dialogue between voice-controlled user interface and user is.

By adjusting an interaction between a voice-controlled user interface and a user is advantageously enabled that the user is not exposed to further mental stress in stressful situations in traffic, as in conventional voice-controlled

Operating interfaces that do not adapt to the mental strain of the user.

It is of particular advantage that a mental stress situation of a user of a voice-controlled user interface in the

predetermined future time interval is detected when the accumulated load value at the expected position and / or in the predetermined future time interval exceeds a predetermined threshold. An adaptation of an interaction between a voice-driven

User interface and a user can therefore already be done

or be prepared before a mental stress situation occurs. This results in the advantage that there is a sufficient time interval between the mental stress situation and the interaction between the voice-controlled user interface and the user, so that the burden on the user in the stress situation is greatly reduced.

It is also advantageous that the adaptation of an interaction in a time shift and / or a suspension and / or a cancellation of a voice input and / or a voice output and / or an overall dialogue between voice-controlled user interface and user is.

In other words, an adaptation of the timing of voice input and / or voice output occurs, thereby reducing the distraction of the user can be. The user can continue to use the voice-controlled user interface if the situation, in particular the load situation, has been simplified.

It may preferably be provided that a current mental stress situation of a user is detected in a method step.

The recognition of a current mental stress situation is advantageous in rapidly changing traffic situations in which the stress situation can not be detected in advance in the context of the procedure.

It is preferably provided that a static load source is an intersection and / or a turning lane and / or a place with a high accident rate and / or a place with speed reductions and / or a place with danger signs.

Furthermore, it can preferably be provided that a dynamic load source is or is a traffic jam and / or construction work and / or weather conditions.

The process is particularly preferably carried out while the

voice-controlled user interface is active. More preferably, it is provided that the current traffic situation and the associated

Possible mental strains are recorded.

Preferably, it is further provided that the minimum Voraussetzun

Procedure of the procedure Measurement data are about: current position, as well as b) the current speed of a motor vehicle.

The current position and / or the current speed can be detected via a vehicle data interface. Further, a preferred

Data model of a road network, in particular a road map, via a navigation component or a navigation module, such as

For example, a navigation system provided. It can the current Position and / or the current speed alternatively or additionally determined by the navigation module. The required information is usually available in the case of a navigation module or a navigation system. In contrast to the prior art, this information is used to estimate or anticipate the current and future mental load of the user in relation to the traffic situation.

It is preferably provided that the at least one load value is adjusted by using user input.

Advantageously, the method thus provides a learning operation or a learning or an optimization phase.

Of course, different users also react differently to stress and mental stress in traffic. Therefore, it is advantageous if the system adapts to the respective user. This adaptation can be done by optimizing the stress values from different stress sources. For this purpose, the user can in particular via the voice-controlled user interface at any time, possibly with the exception of a time in which the voice input was moved, give feedback or feedback about how he is the current or

experienced previous load situation.

Depending on whether this is a positive concrete feedback

(for example, the spoken feedback: "this crossing was not bad") or negative concrete feedback (for example the

spoken feedback: "This lane change was very difficult"), the load value for the load situation, to which the

Feedback relates, decreased or increased.

It is further preferred that at least one load source is a general type-specific load value and / or a personal type-specific load value and / or a more general concrete one Load value and / or a personal concrete load value is assigned.

Specific load values refer to a specific event, such as a particular intersection. In addition, specific stress values can be assigned to local places such as hospitals, schools, kindergartens or nursing homes. By contrast, type-specific loads apply to entire categories, such as intersections in general. General load values are values that are given to the method or that accompany a device that performs the method. General load values may be dynamically reloaded during the performance of the method or use of the device. Furthermore, general load values are not specifically assigned to a user, but can be calculated, for example, by a notification of load values of multiple users. Personal values are directly to the

User assigned, which can directly influence this value. All of these data can advantageously be stored in a database. Other events that are a general type-specific

Load value and / or a personal type-specific load value and / or a general concrete load value and / or a personal concrete load value are the local or general traffic, for example in the so-called "rush hour" or the evening rush, or the passenger and public In particular with regard to local places such as hospitals, schools, kindergartens or retirement homes, the passenger traffic can lead to specific, type-specific, general or personal burdens, so that corresponding load values can be assigned.

It is further preferred that route information of a navigation module be taken into account for determining the expected position.

By using route information, in particular when a motor vehicle driver as the user of the voice-controlled user interface follows the route information, the accuracy of the determination of a expected position of the motor vehicle can be specified within a predetermined future time interval.

In a further preferred embodiment it is provided that

Loading sources and / or load values are updated, the update by consulting an external database, in particular an external database of a cloud system and / or a server system, and / or a traffic message channel (TMC) is carried out.

A traffic message channel is an additional information channel in addition to the local radio program (TMC standard ISO 14819).

Furthermore, it can preferably be provided that load values are below

Considering health values or health data of a user of a voice-controlled user interface to be adjusted.

For example, an adjustment of the load values under

Taking into account the age of the motor vehicle driver. Since older road users often experience situations as stressful or at a higher stress level than younger road users, the general, personal, concrete and / or type-specific load values can be increased taking into account the age of the motor vehicle driver so that stress situations are detected more frequently for an older road user than for a younger motor vehicle driver. Older motor vehicle drivers will be further relieved.

Another object of the invention is to provide a device suitable for carrying out the invention

method described above, comprising a voice-controlled

Operator interface, a navigation module, a database

Vehicle data module, and a load evaluation module, wherein the

Vehicle data module is adapted to determine at least the position and / or the speed of a motor vehicle, wherein the navigation module is adapted to provide a data model of a road network in an environment of a position of a motor vehicle, wherein the Load evaluation module is formed, for a future predetermined time interval an expected position of a motor vehicle under

Use at least the position and speed of the

Motor vehicle and the data model of the road network to be determined, it being further provided that the database is adapted to store static and / or dynamic load sources, and that the

Stress evaluation module is formed, a mental stress situation of a user of the voice-controlled user interface in the

to recognize predetermined future time interval.

Preferably, the load evaluation module is designed to link load sources with data points of the data model of the road network.

Furthermore, the load evaluation module can be designed to assign at least one load source at least one load value.

More preferably, the load evaluation module is designed to determine an accumulated load value from the at least one load value and / or at least one load value as accumulated

Take over load value.

More preferably, the stress evaluation module is configured to detect the mental stress situation if the accumulated stress value at the expected position and / or in the predetermined future time interval exceeds a predetermined threshold value.

Furthermore, it can preferably be provided that the load evaluation module is designed to have an interaction between the voice-controlled module

Control interface and a user, if one

Strain situation was detected, the adjustment is particularly preferably a time shift and / or suspension and / or cancellation of a voice input and / or a voice output and / or a total dialogue between the voice-controlled user interface and the user. Preferably, the load evaluation module is designed to adjust load values using user input.

This advantageously allows a learning operation or a learning or optimization phase.

Furthermore, it can preferably be provided that a traffic message channel data source is provided, wherein the database is designed to receive data about load sources from the traffic message channel data source.

This advantageously achieves that current traffic information flows into the recognition of a mental stress situation.

Further, preferably a cloud system and / or a server system

be provided, the database is designed to transfer data about

Receive pollution sources from the cloud system and / or the server system.

By providing a cloud or a server system or an external database created therein, it becomes possible to include load values, in particular general load values, over several users.

Furthermore, it can preferably be provided that the load evaluation module is designed to determine the expected position

Route information of the navigation module to be considered.

Another solution to the problem is to provide a

Motor vehicle with a device described above.

Brief description of the figures

An embodiment of the present invention will be illustrated with reference to the figures. Show it: Fig. 1 is a flowchart of a method for controlling a

voice-controlled user interface in a motor vehicle, and

Fig. 2 is a schematic representation of an apparatus for performing the method.

Detailed description of the figures

Fig. 1 shows a flowchart of a method for controlling

voice-controlled operating interfaces in a motor vehicle. In a first step Sl of the method becomes an expected position of a

Motor vehicle within a predetermined future time interval T determined. In a following step S2, static and / or dynamic load sources are determined at the expected position. In step S3, at least one load value is assigned to at least one of the determined static and / or dynamic load sources. In the subsequent method step S4, an accumulated load value is determined from the at least one load value and / or at least one load value is adopted as an accumulated load value. In method step S5 following method step S4, a mental loading situation of a user of a voice-controlled operator interface is detected in the predetermined future time interval if the accumulated load value at the expected position and / or in the predetermined future time interval T exceeds a predetermined threshold value. The process then proceeds to step S6, in which an interaction between a

voice-controlled user interface and a user is adapted when a load situation has been detected, wherein the adjustment is a time shift and / or suspension and / or termination of a voice input and / or a voice output and / or an overall dialogue between voice-controlled user interface and user. However, except for a time shift and / or suspension and / or termination of a voice input and / or voice output and / or an overall dialogue, TMC hazard messages and user interaction may be included in a hazard message, such as the Indication of a driver coming on the same lane, a so-called "ghost driver".

Thus, exceptions can be made to the time shift and / or suspension and / or termination of a voice input and / or a

Speech output and / or a complete dialog can be specified or defined.

FIG. 2 shows a device 100 suitable for carrying out the method illustrated in FIG. The device 100 consists of a client system

10 and a server or cloud system 11. The client system 10 includes a voice-controlled user interface 12, which to a

Speech recognizer 13 and a voice output 14 is coupled. In addition, in the client system 10 there is a load evaluation module 15, a navigation module 16, a vehicle data module 17, a traffic message module.

Channel data source 18, a database 19 containing possible user-specific stress sources and their stress value. This database 19 is synchronized with the cloud system 11 via a user management and synchronization component 20. For this purpose, the cloud system 11 comprises an external database 21 with user-spanning or general user

Load sources and load values kept up to date by internal services 22, which in turn access external services 23. The load evaluation module 15 is designed to adapt an interaction of the voice-controlled user interface 12 with a user. The connection to the server or cloud system 11 is optional and the method or device also works based on the information stored in the local database 19. The vehicle data module 17 and / or the navigation module 16 provide the

Load evaluation module 15 Measurement data on the current position and the current speed of the motor vehicle available. Furthermore, the navigation module 16 provides the load evaluation module 15 with a data model of a road network in an environment of a position of a motor vehicle, in particular a road map. Within the load evaluation module 15, a model is calculated which characterizes or determines the position of the vehicle within a future predetermined time interval T to be parameterized. This model refers to the current position and the current speed of the

Motor vehicle with a to all the possibilities for the position within the

Time window or the time interval T to calculate. The model, in particular the position model, can be improved in the event that the user follows a previously entered route from the navigation component or the navigation module 16. The following will be

Positional model annotated with potential mental stress for the user.

Such loads include:

Intersections on the map, especially if you want to turn left based on the previously entered route, sharp turns on the map, planned or possible lane change based on the previously entered route and the road map, traffic signs with danger symbols or abrupt speed reductions, the traffic signs for example via Internet services, the road map or cameras are detected, dense traffic reported via internet-based systems or via local cameras in the

Background working image processing systems, traffic jams, accident or

Traffic obstruction messages via the traffic message channel, and accident statistics, such as from bends or intersections, hospitals, schools, kindergartens or retirement homes, the local or general

Traffic, passenger transport, local public transport.

At least one of the following load values is assigned to these possible load sources: general type-specific load value, personal type-specific load value, more generally more concrete

Debit value or personal specific debit value.

The load values to be allocated are located, for example, in the database 19. The database 19 can be enriched directly via the traffic message channel data source 18 with current traffic data.

In addition, the database 19 with the external database 21 of the server or cloud system 11 via the user management and Synchronization component 20 synchronizes and thus also receives external, current and complicated to be calculated load sources via internal services 22, which use external services 23.

The stress evaluation module 15 determines an accumulated one

Load value. This accumulated load value is calculated from the aforementioned possible load values, for example, via a prioritization or by means of a linear combination of the load values. This process takes place on a regular basis to ensure that the

voice-controlled user interface 12 has always up-to-date information about the load status of the user.

The load evaluation module 15 transmits the accumulated

Load value to the voice-controlled operator interface 12, which can respond as follows to the user before additional load

Protect voice interaction. If necessary, speech output to be outputted and pending through the speech output 14 may be interrupted and delayed in time as soon as a user-specific threshold value is exceeded by the accumulated debit value.

In this case, from a time shift and / or suspension and / or termination of a voice input and / or a voice output and / or an overall dialogue TMC danger messages and interactions of the user may be excluded as part of a danger message, such as the reference to a on the same lane

Coming driver, a so-called "ghost driver".

Speech output and / or a complete dialog can be specified or defined.

Alternatively, the voice output or the entire dialogue, possibly with a very short notification or a signal, also be canceled if the load, in particular the load value is too large. In addition, the user for pending voice input via the

Speech recognizers 13 get more time, so he earliest directly in the

Connection to the source of stress or the load situation must respond. Alternatively, the voice input or the entire dialogue, possibly with a very short notification or a signal, also be canceled if the burden or the

Load value is too large.

To optimize the load values, the user can, for example, via the voice-controlled user interface 12 at any time, possibly up to a time in which the voice input has been moved,

Give feedback on how he experienced the current stress situation. The feedback is received by the load evaluation module 15. Depending on whether this is a concrete positive feedback or a concrete negative feedback, the load value for the corresponding load source, ie for the corresponding annotation of the position model or a data point of the data model of the road network, is lowered or increased. Also, an annotation of a position closest to the current position and type may either be lowered or increased. The same applies to type-specific

Stress values, for example, based on spoken user input such as "crossings are not a problem" as opposed to "I hate lane change". In principle, the user-specific threshold values for the interruption / shifting of the voice input and voice output or the interruption of the dialog can also be adapted. Adapted load values are stored in the database 19. If the client system 10 of the device 100 is connected to a server or cloud system 11, then these adjusted load values arrive via the user management and synchronization component 20 in the external database 21 of the cloud or server system 11 Service within the internal services 22 takes care of adjusting the general concrete load values based on the personal concrete load values across all users of the system. In addition, the averaged values are aggregated via the personal type-specific load values to the general type-specific load values.

Claims

claims

1. A method for controlling voice-controlled operating interfaces (12) in motor vehicles comprising the steps of: a) determining an expected position of a motor vehicle within a predetermined future time interval (T), b) determining static and / or dynamic

Load sources at the expected position, c) Assigning at least one load value to at least one of the determined static and / or dynamic

D) determining an accumulated load value from the at least one load value and / or taking at least one load value as the accumulated load value, e) detecting a mental load situation of a user of a voice-operated operator interface (12) in the predetermined future time interval (T), wherein the mental Stress situation is detected when the accumulated load value at the expected position and / or in the predetermined future time interval exceeds a predetermined threshold, and f) adjusting an interaction between a

speech-controlled user interface (12) and a user when a load situation has been detected, the adjustment being a time shift and / or a suspension and / or a cancellation of a voice input and / or a voice output and / or a Overall dialog between voice-activated operator interface (12) and user is.

A method according to claim 1, characterized in that the at least one load value with the involvement of

User input is adjusted.

A method according to claim 1 or 2, characterized in that at least one load source, a general type-specific load value and / or a personal type-specific

Load value and / or a general concrete load value and / or a personal concrete load value is assigned.

Method according to one of the preceding claims, characterized in that for determining the expected position route information of a navigation module (16) are taken into account.

Method according to one of the preceding claims, characterized in that load sources and / or load values are updated, wherein the update by consulting an external database (21), in particular an external database (21) of a cloud system (11) and / or a server -System, and / or a traffic message channel.

Device (100) suitable for carrying out the method according to one of claims 1 to 5, comprising a voice-controlled user interface (12), a navigation module (16), a database (19), a vehicle data module (17), and a

Load evaluation module (15), wherein the vehicle data module (17) is adapted to determine at least the position and / or the speed of a motor vehicle, wherein the navigation module (16) is adapted to a data model of a road network in one

Provide environment of a position of a motor vehicle, wherein the load evaluation module (15) is formed, for a future predetermined time interval an expected position of a

Motor vehicle using at least the position and the speed of the motor vehicle and the data model of the road network to determine, wherein the database (19) is adapted to store static and / or dynamic load sources, and wherein the load evaluation module (15) is formed, a mental load situation of User of the voice-controlled user interface (12) to recognize in the predetermined future time interval.

Device (100) according to claim 6, wherein the

Load evaluation module (15) is adapted to link load sources with data points of the data model of the road network, and / or wherein the load evaluation module (15) is adapted to assign at least one load source at least one load value.

Device (100) according to claim 7, wherein the

Load evaluation module (15) is adapted to determine an accumulated load value from the at least one load value and / or at least one load value as accumulated

Take over load value.

Apparatus (100) according to claim 8, wherein said

Load evaluation module (15) is formed, the mental

Recognize load situation when the accumulated load value at the expected position and / or in the predetermined future time interval a predetermined threshold

exceeds.

Device (100) according to one of claims 6 to 9, wherein the

Load evaluation module (15) is adapted to adjust an interaction between the voice-controlled user interface (12) and a user when a load situation has been detected, the adjustment preferably a time shift and / or is a suspension and / or an abort of a voice input and / or a voice output and / or an overall dialogue between the voice-operated operator interface (12) and user.

11. Device (100) according to one of claims 6 to 10, characterized

in that the load evaluation module (15) is designed to adjust load values using user inputs.

12. Device (100) according to any one of claims 6 to 11, characterized

characterized in that a traffic message channel data source (18) is provided, wherein the database (19) is adapted to receive data about stress sources from the traffic message channel data source (18).

13. Device (100) according to one of claims 6 to 12, characterized

characterized in that a cloud system (11) and / or a server system is provided, wherein the database (19) is adapted to receive data on load sources from the cloud system (11) and / or the server system.

14. Motor vehicle with a device (100) according to one of claims 6 to 13.